The increase in cohabitation is one of the most significant shifts in family demographics of the past century (Casper &Bianchi, 2001;Manning &Smock, 2002;Smock, 2000). The significance of this increase is that, contrary to the expectations of many, cohabitation is associated with increased risk for marital distress and divorce (e.g., Kamp Dush, Cohan, &Amato, 2003;Stanley, Whitton, &Markman, 2004)-the so called "cohabitation effect." Current speculation for the cohabitation effect suggests theories related to selection (i.e., people who choose to cohabit have pre-existing sociodemographic risks for poor outcomes) and experience (i.e., something about cohabitation itself makes poor outcomes more likely). To date, methodological limitations of previous studies (e.g., lack of theory, retrospective accounts, weak measurement, small samples, infrequent follow-ups, and data from only one partner) have precluded an understanding of just what the experience of cohabitation is and how it is linked to higher risk factors for divorce. Without understanding mechanisms of risk, the most effective modes of lowering such risks cannot be developed. This project is designed to advance understandings of cohabitation effect by examining couple development using the overarching framework of commitment theory (e.g., Stanley &Markman, 1992). The proposed project will be the first comprehensive longitudinal study focused on cohabitation and it will begin at an earlier stage of couple development than any other study on cohabitation. Specifically, the proposed study aims to (a) Examine the roles of selection and experience in explaining the cohabitation effect by comprehensively measuring both in a longitudinal, nationally representative sample of dating and cohabiting individuals (N = 1200), (b) Test hypotheses framed in commitment theory that can explain why the experience of cohabitation is associated with poor relationship outcomes for some couples, (c) Use data from a sub-sample that includes both partners in couples to examine how discrepancies in commitment relate to the cohabitation effect, (d) Expand the foundation of knowledge upon which interventions designed to lower risks and raise protective factors for dating and cohabiting couples can be built. Random digit dialing techniques will be used to identify the sample and participants will complete questionnaires by mail every four months.